Need examples of acceptable/unacceptable derivative images

← Return to forum

  • Hello -

    I was directed here by my university librarian, I hope you don't mind if I ask a question, as follows:

    The scientists I work with make images for their journal articles, say of a particular protein. Later on, they may want to use those same images in another venue - a review chapter, another journal article, etc.

    They've somehow developed the notion that if they change their original image modestly - say, color a section differently or rotate the image, that they then have produced an image 'new' enough to avoid the copyright restrictions held by the original journal. I've tried to communicate to them that small changes aren't sufficient, but I think visual examples would help them understand.

    Can anyone point me to a website or text or other resource that shows examples of acceptable and unacceptable derivative images?

    Thanks in advance,
    D. B.
  • Am I correct in assuming the reason for this question is that the scientists in question retain the right to make derivative works when they sign over rights to their articles for publication?

    If that's the case, I'm not sure that this is a question of copyright law. Instead, I think that it would depend upon the contract language.
  • [quote=Freya Anderson]Am I correct in assuming the reason for this question is that the scientists in question retain the right to make derivative works when they sign over rights to their articles for publication?

    If that's the case, I'm not sure that this is a question of copyright law. Instead, I think that it would depend upon the contract language.[/quote]
    Thank you for your reply.

    Unfortunately there isn't a consistent answer regarding contract language - some publishers are fairly open to letting the scientists re-use their work, even allowing internet postings, while other publishers are quite restrictive, wanting for example use of an image restricted to classroom use unless permission is granted. The mixed message has I think encouraged the 'a few changes makes a new image' attitude by scientists, who don't want to spend time researching their publication contracts or making completely new images.

    So, I'm trying to clarify for them what a derivative image is, and how much work needs to be done to a pre-existing image before it can be defined as 'new' in a way that won't cause any copyright concerns.

    D. B.
  • I agree that color changes and rotations are trivial changes that do not create a derivative image. However, I question the need for derivative images.

    I believe that many publishers do not understand copyright law, particularly fair use. Many publishers insist that no use is fair unless permission is granted. For example, consider this completely false statement from the Business Software Alliance: "It is illegal to copy software, music, films, video games, or any other form of intellectual propert without the copyright holder's permission. In fact, it's stealing." Top of page 11

    I have never seen a contract that explicitly signs away image rights. I have only seen this issue discussed in documents like the BSA one which clearly misrepresents (perhaps unintentionally) the true situation. Unless I see the terms from an actual contract, I assume that fair use has not been signed away.

    Assuming that fair use has not been signed away, here's my fair use analysis. By the way, I am working on articles with images. Although none have been published yet, I am a producer of copyrighted material similar to what is being discussed.

    Four factors (Anyone wishing to see a more detailed explanation of the four factors can find numerous examples on this forum):
    1. Educational. For fair use.
    2. Factual in my case and in the case of most scientific images that I have seen. For.
    3. Small %. For. Although I think publishers disagree, I view an article with images as a whole with many parts, not a collection of wholes.
    4. No market effect. For. I don't think any scientist who creates images that are marketable on their own would give them away with an article.

    However, if fair use has indeed been signed away, I agree that trivial changes do not get around this. When I have time, I will look for guidelines on derivative images. However, I have never encountered any that would apply to this situation.
  • A derivative work is a work based on the original. It's easy to understand when we think of the written word - translations, parodies, book to screenplay etc.

    When we think of images, it is not as clear (at least for me).

    I think that modifications to images are derivative works if they are newly created, original works. So if you change the background color, you have not created a new original work. If you distort an image, do you have a brand new original copy or just a work that you have tampered with? There is a point when manipulating an image could produce a new work, but then the question is, is that work based on the original - derivative work - or did the amount of manipulation actually create a new work not at all based on the original.

    So I think that a derivative work must be recognizable as something based on the original. The new work must be original enough to be a new work based on the original with its own copyright protection. If you manipulate a work, you might create a derivative work. But over manipulation reaches a point that an entirely new work is created because it has no relationship to the original work.

    I tend to agree that the scientists should go ahead and use their images unless a contract says otherwise. I would also suggest to the scientists that they slap a copyright notice on future images created by the scientists for publication.
  • Some color changes are more significant than others:
  • I would also suggest that you take every opportunity to educate these scientists to read their contracts carefully before signing. So many authors don't bother to do so! You could encourage them to be proactive and not submit their papers to journals that don't let them retain rights to reuse their own work -- or at least to try to revise clauses in their contracts that restrict the uses they might want to make in the future. As an author, that's something I always check for before signing, and as an editor, I've made sure my journal's contract states clearly that any author can reuse their work as long as they acknowledge its initial publication in our journal.

    I would suggest a useful rule of thumb for "derivative images" is: has value been added? In the case of the Warhol print linked above, value has arguably been added to the original image. Did Duchamp add value when he drew a mustache on the Mona Lisa? That's a much grayer area.

Posting to the forum is only available to users who are logged in.

← Return to forum